23

CHAPTER II
OVERVIEW OF STABILITY ANALYSIS AND DESIGN METHODS
IN AISC (2010)
2.1 Background
Chapter C of AISC (2010), Design for Stability, states that any analysis and
design procedure that addresses the following effects on the overall stability of the
structure and its elements is permitted:
1. Flexural, axial and shear deformations (in members, connections and other
components), and all other deformations that contribute to displacements of the
structure,
2. Reduction in stiffness (and corresponding increases in deformations) due to residual
stresses and material yielding,
3. P- (P-large delta) effects, which are the effects of axial loads P acting through the
relative transverse displacements of the member ends (see Fig. 2.1),
4. P- (P-small delta) effects, which for initially straight members loaded by bending in
a single plane, are due to the member axial load acting through the transverse bending
displacements relative to the member chord (see Fig. 2.1),
5. P-
o
and P-
o
effects, which are caused by the member axial loads acting through
unavoidable initial
o
and
o
geometric imperfections (within fabrication and erection
tolerances) relative to the ideal configuration of the structure, and
6. Uncertainty in stiffness and strength.
The above statement gives the Engineer the freedom to select or devise methods that are
best suited for the many different routine and nonroutine situations encountered in
24

practice. It allows for innovation within the constraints of the proper consideration of the
physical effects that influence the structural response.

Figure 2.1. Second-order P- and P- effects.
Since the P- and P- effects are central components of the frame stability
behavior, it is useful to elaborate on their definitions. As illustrated in Figure 2.1, any
relative transverse displacement between a members ends produces a couple of P
times , where P is the axial force transmitted by the member. This couple must be
resisted by the structure. In typical tiered building systems, the predominant P- effects
come from the vertical columns. However, clear-span gable frames also have a P-
effect associated with the axial thrust in the rafters and the relative transverse
displacement between the ends of the rafter segments, as shown in Figure 2.2. In certain
types of structures, e.g., some types of modular frames, the predominant P- effects
come from simply-connected gravity columns (leaner columns), which depend on the
lateral load resisting system for their lateral stability. Figure 2.3 gives a simple
25

illustration of this behavior. Under a sidesway displacement of the structure , a lateral
force equal to P/L is required to maintain equilibrium of the leaner column in the
deflected configuration. This lateral force must be resisted by the structure.


Figure 2.2. P- effects in a rafter segment of a gable frame.

Figure 2.3. Illustration of P- effects from a gravity (leaner) column.
Members that have small transverse displacements relative to their rotated chord
and/or small axial forces have small P- effects. This includes leaner columns, which are
commonly idealized as straight pin-ended struts and therefore have zero and zero P-
effects, as well as stiff columns that deflect in sidesway mainly due to end rotation of the
26

adjacent beams (see Figure 2.4). However, members such as that shown in Figure 2.1
must resist additional moments of P times at the various cross-sections along their
lengths. These P- moments increase the member deformations, and therefore they
reduce the net member stiffnesses and increase the net sidesway displacements .
Interestingly, if the structure is subdivided into a large number of short-length elements,
the representation of the P- effect in each element is sufficient to capture both the
overall member P- and the internal member P- effects. Figure 2.2 is to some extent
indicative of this attribute.

Figure 2.4. Illustration of deformed geometry resulting in small P- effects.
Guney and White (2007) study the number of elements required per member for
P-large delta only analysis procedures to ensure less than 5 % error in the nodal
displacements and less than 3 % error in the maximum internal moments for second-
order elastic analysis of prismatic members with a wide range of loadings and end
conditions. They also address the number of elements required to ensure less than 2 %
27

error in eigenvalue buckling analysis solutions. Based on the research by Guney and
White (2007), the AISC Design Guide 25 provides tables showing the required number of
elements to achieve the desired analysis accuracy for a given calculated P
r
/P
eL
or P
r

/
eL
P value for sway columns with simply-supported bases, sway columns with top and
bottom rotational restraints, and rafters and non-sway columns, where P
eL
is the member
elastic buckling load based on the idealized simply-supported end conditions and nominal
elastic stiffness and
eL
P

is the member elastic buckling load based on the idealized
simply-supported end conditions and reduced elastic stiffness specified by the direct
analysis method. For example, at P
r
/P
eL
= 0.15, three elements are required for a sway
column to ensure less than 5 % error in the nodal displacements and less than 3 % error in
the maximum internal moments. In general, P-large delta only analysis procedures can
adequately capture internal member P- effects when the subdivisions of members
achieve P
r
< 0.02P
e"
or

P
r
< 0.02
" e
P , where P
e"
is the element elastic buckling load
based on idealized simply-supported end conditions and nominal stiffness and
" e
P element
elastic buckling load based on idealized simply-supported end conditions and reduced
stiffness specified in the direct analysis method. Second-order analysis methods that
directly include both P- and P- effects at the element level generally provide better
accuracy than P-large delta analysis procedures.
In tapered-web and general nonprismatic members, the centroidal axis is not
straight, thus causing additional moments of P times y, where y is the shift in the
centroidal axis relative to a straight chord between the cross-section centroids at the
member ends (see Figure 2.5). This important effect is incorporated within a proper first-
28

order analysis, by virtue of the correct modeling of the geometry. Also, this is a member
initial curvature effect rather than a P- effect. Additional P- moments are caused by
the transverse bending displacements associated with the primary moments P times y.
Typically, the initial curvature effect is incorporated in part by using multiple elements
along the member length and locating the nodes of the analysis model along the curved
cross-section centroidal axis. However, when one or both cross-sections are singly-
symmetric, there is an abrupt shift in the centroidal axis at cross-section transitions.
Also, it is convenient to use a straight reference axis that has a variable offset from the
centroidal axis in some situations (e.g., placing the reference axis at a constant depth
below the top of the steel in the rafters). In these cases, the first moment of the cross-
sectional area is non-zero with respect to the reference axis.

Figure 2.5. Member initial curvature effect of P times y.
Since the inception of the 1961 edition of the AISC Specification, when the
concept of column effective length was first introduced by AISC, American design
procedures generally have addressed all of the above effects in some fashion whenever
they were deemed to have an important influence on the structural response. Member
yielding, residual stress effects, and geometric imperfection effects traditionally have
been addressed in the formulation of member design resistances, and have not been
29

considered in the analysis except the following case. Engineers have often included a
nominal out-of-plumbness effect in the analysis of gravity load combinations, particularly
if the geometry and loading are symmetric. Strictly, this is not necessary for the in-plane
strength assessment of beam-columns in the prior AISC Specifications. However, this
practice is necessary to determine P-
o
effects on bracing forces, beam moments,
connection moments, and in-plane member moments for checking the out-of-plane
resistance of beam-columns. There also has always been implicit recognition that the
engineers can use their professional judgment to disregard specific effects (e.g., member
shear deformations, connection deformations, etc.) whenever they are deemed to be
negligible.
Furthermore, the 1961 AISC Specification, and other AISC Specifications up
until 1986, relied strictly on the structural analysis only for calculation of linear elastic
forces and moments within the idealized perfectly straight and plumb nominally elastic
structural system. The influence of second-order (P- and P-) effects was addressed
solely by an amplifier applied discreetly to the flexural stresses from the linear elastic
analysis, via the following beam-column strength interaction equation:
0 . 1
1

+
b
e
a
b m
a
a
F
F
f
f C
F
f
(Eq. 2.1)
The expression
AF
F
f
C
e
a
m
=

1
(Eq. 2.2)
30

in Eq. 2.1 is an approximate amplification factor (AF) for the member flexural stresses f
b
,
accounting for the second-order P- and P- effects. LeMessurier (1977) and others later
addressed the specifics of how to properly determine these amplified bending stresses in
general cases, including multi-story structures and single-story structures with significant
leaner column loads, all within the context of Allowable Stress Design. However, when
the AISC (1989) ASD provisions are used, Engineers often apply Eq. 2.1 in ways that can
significantly underestimate the physical second-order effects in certain types of
structures. The accuracy of the calculations hinges largely on the proper determination of
F
e
, the member axial stress at incipient elastic buckling (considering the interaction of the
member with the rest of the structure). The parameter
e
F is calculated by dividing this
buckling stress by the column factor of safety, 23/12 = 1.92. The axial stress
e
F is
determined typically using the equation
2
2
) / ( 23
12
b b
e
r KL
E
F

= (Eq. 2.3)
where KL
b
/r
b
is the column effective slenderness ratio in the plane of bending, and K is
the effective length factor associated with the above buckling solution. Also, this K
factor is used typically in the calculation of the column axial resistance F
a
. If desired,
e
F = F
e
/1.92 can be determined directly from the elastic buckling analysis model.
Obviously if the correct
e
F is large relative to f
a
, the second-order P- and P- effects
are small. The term C
m
in Eqs. 2.1 and 2.2 is discussed subsequently.
AISC (1986) LRFD was the first American Specification to refer explicitly to the
calculation of second-order moments from a structural analysis. This specification
31

introduced the following two-equation format for its primary beam-column strength
interaction curve:
0 . 1
2

n b
u
n c
u
M
M
P
P
for 2 . 0 <

n c
u
P
P
(Eq. 2.4a)
0 . 1
9
8

n b
u
n c
u
M
M
P
P
for 2 . 0

n c
u
P
P
(Eq. 2.4b)
where M
u
is defined as the maximum second-order elastic moment along the member
length. AISC (1986) states that M
u
may be determined from a second-order elastic
analysis using factored loads. However it also provides an amplification factor procedure
for calculation of the second-order elastic moments from a first-order elastic analysis.
This procedure is in essence an approximate second-order analysis. The above moments
M
u
are the second-order elastic moments in the idealized initially plumb and straight,
nominally elastic structure.
In all the AISC Specifications from AISC (1961) through AISC (1989) ASD and
AISC (1999) LRFD, the influence of geometric imperfections and residual stresses was
addressed solely within the calculation of the member resistances (F
a
and F
b
in ASD and
P
n
and M
n
in LRFD). The direct analysis provisions that is first introduced in AISC
(2005) and addressed in Chapter C of AISC (2010) recognize that specific advantages
can be realized by moving an appropriate nominal consideration of these effects out of
the resistance side and into the structural analysis side of the design equations. By
incorporating an appropriate nominal consideration of these effects in the analysis, the
resistance side of the design equations is greatly simplified and the accuracy of the design
procedure is generally improved. These attributes are discussed further in the subsequent
sections.
32

It is important to note that all of the above design procedures are based inherently
on the use of second-order elastic analysis (first-order elastic analysis with amplifiers
being considered as one type of second-order elastic analysis). Also, one must recognize
that elastic analysis generally does not include the consideration of the member
resistances in itself. Therefore, all of the above methods must include member resistance
equations. However, the method of analysis and the equations for checking the member
resistances are inextricably linked. Changes in the analysis calculation of the required
strengths, e.g., f
a
and f
b
C
m
/(1-f
a
/F'
e
) in Eq. 2.1 or P
u
and M
u
in Eqs. 2.4, can lead to
simplifications in the member resistances, typically F
a
in Eq. 2.1 or P
n
in Eqs. 2.4.
Specifically, if the structural analysis can be configured to provide an appropriate
representation of the internal member forces, the in-plane resistance of the structure can
be checked entirely on a cross-section by cross-section basis. This is discussed in Section
2.2.
AISC (2005 and 2010) adopts the equations from AISC (1999) LRFD as a base
representation of the beam-column resistances for all of its analysis and design
procedures. However, the notation is generalized such that the equations apply to both
ASD and LRFD:
0 . 1
2
+
c
r
c
r
M
M
P
P
for 2 . 0 <
c
r
P
P
(Eq. 2.5a, AISC H1-1a)


0 . 1
9
8
+
c
r
c
r
M
M
P
P
for 2 . 0
c
r
P
P
(Eq. 2.5b, AISC H1-1b)
The terms in these equations are defined as follows:
P
r
= the required compression strength, determined in ASD by analyzing the structure
under 1.6 times the ASD load combinations and then dividing the results by 1.6, or
33

determined in LRFD by analyzing the structure under the LRFD load combinations.
M
r
= the required flexural strength, determined in ASD by analyzing the structure under
1.6 times the ASD load combinations and then dividing the results by 1.6, or determined
in LRFD by analyzing the structure under the LRFD load combinations.
P
c
= the allowable or design compression resistance, given by P
n
/
c
in ASD or by
c
P
n
in
LRFD, where P
n
is the nominal compression resistance determined in accordance with
Chapter E.
M
c
= the allowable or design flexural resistance, given by M
n
/
b
in ASD or by
b
M
n
in
LRFD, where M
n
is the nominal flexural resistance determined in accordance with
Chapter F.

c
and
b
= resistance factors for axial compression and bending, both equal to 0.9.

c
and
b
= factors of safety for axial compression and bending, both equal to 1.67.
For Equations 2.5a and b, another equation numbers are shown. In this dissertation,
AISC (2010) equation numbers are denoted by AISC followed by the equation number.
In many cases, Equations 2.5a and b provide a more liberal characterization of the beam-
column resistances than the multiple beam-column strength curves in AISC ASD (1989).
It should be noted that the 1.6 factor applied to determine the required
compression and flexural strengths under the ASD load combinations is smaller than the
column safety factor of 1.92 within the AISC ASD (1989) amplification of the flexural
stresses (see Eqs. 2.1 through 2.3). However, ASD-H1 also states that C
m
shall be taken
as 0.85 in Eqs. 2.1 and 2.2 for frames subject to joint translation. This C
m
value typically
underestimates the sidesway moment amplification effects (Salmon and Johnson 1996).
Nevertheless, the ASD moment amplifier summarized in Eq. 2.2 is still conservative in
34

many practical cases. This is because the predominant second-order effects are often
associated solely with the structure sidesway. Equation 2.1 applies a single amplifier
indiscriminately to the total flexural stresses from both non-sway and sidesway
displacements. The amplification factor procedure in AISC (1999) LRFD and AISC
(2005 and 2010) is more accurate, but involves a cumbersome subdivision of the analysis
into separate no-translation (nt) and lateral translation (lt) parts. Kuchenbecker et al.
(2004) and White et al. (2007a & b) outline an amplified first-order elastic analysis
approach that provides good accuracy for rectangular framing. This approach avoids the
above cumbersome attributes of the AISC (1999, 2005 & 2010) amplification factor
procedure
In the subsequent developments, it is useful to consider the characterization of
separate in-plane and out-of-plane beam-column resistances using Eqs. 2.5. The in-plane
beam-column resistance is addressed by neglecting out-of-plane flexural and/or flexural-
torsional buckling in the calculation of P
c
and by neglecting lateral-torsional buckling in
the calculation of M
c
. Correspondingly, the out-of-plane resistance is defined by
considering solely the out-of-plane flexural and flexural-torsional buckling limit states in
the determination of P
c
(excluding the in-plane flexural buckling limit state), and by
considering all the potential flexural limit states (lateral-torsional buckling, flange local
buckling, tension flange yielding and general yielding) in the calculation of M
c
. The
interaction of general yielding and/or local buckling with column flexural buckling is
included inherently within the calculation of both the in-plane and out-of-plane axial
strengths P
ci
and P
co
. However, in the AISC (2005 and 2010) flexural resistance
equations, lateral-torsional buckling, compression flange local buckling and tension
35

flange yielding are handled as separate and independent limit states. The smaller
resistance from these separate limit states generally governs the flexural resistance. This
research recommends that the compression flange local buckling and tension flange
yielding limit states should be included in the calculation of the anchor point M
co
for the
beam-column out-of-plane strength using Eqs. 2.5. Otherwise, potential influences of
compression flange local buckling or tension flange yielding on the out-of-plane beam-
column resistance are neglected.
2.2 Direct Analysis Method
Table 2.1 summarizes three specific overriding stability design procedures
defined in AISC (2010):
1. The direct analysis method, detailed in Sections C2 and C3,
2. The effective length method, detailed in Appendix 7, and
3. The first-order analysis method, detailed in Appendix 7.
Within the restrictions specified on their usage, and provided that effects such as
connection rotations or member axial and shear deformations are properly considered in
the analysis when these attributes are important, each of these methods is intended to
comprehensively address all of the effects listed in the beginning of Section 2.1.
As seen in Table 2.1, the direct analysis method is the only one of the above three
procedures that is generally applicable. In basic terms, this method involves the
following simple modifications to the second-order elastic analysis: (1) the use of a
reduced elastic stiffness and (2) for rectangular or tiered structures, the use of a notional
lateral load equal to a fraction of the vertical load at each level of the structure. These
two devices are adjustments to the second-order analysis that account for: (1) member
3
6


T
a
b
l
e

2
.
1
.


S
u
m
m
a
r
y

o
f

A
I
S
C

(
2
0
1
0
)

p
r
o
v
i
s
i
o
n
s

f
o
r

s
t
a
b
i
l
i
t
y

a
n
a
l
y
s
i
s

a
n
d

d
e
s
i
g
n
.


D
i
r
e
c
t

A
n
a
l
y
s
i
s

(
S
e
c
t
i
o
n
s

C
2

a
n
d

C
3
)

E
f
f
e
c
t
i
v
e

L
e
n
g
t
h

(
A
p
p
e
n
d
i
x

7
)

1
s
t
-
O
r
d
e
r

A
n
a
l
y
s
i
s

(
A
p
p
e
n
d
i
x

7
)

L
i
m
i
t
a
t
i
o
n
s

o
n

t
h
e

U
s
e

o
f

t
h
e

M
e
t
h
o
d

N
o
n
e

2
n
d
/

1
s
t

<

1
.
5

2
n
d
/

1
s
t

<

1
.
5
,

P
r

<

0
.
5

P
y

(
S
e
e

N
o
t
e

6
)

T
y
p
e

o
f

A
n
a
l
y
s
i
s


2
n
d
-
o
r
d
e
r


(
S
e
e

N
o
t
e

1
)

2
n
d
-
o
r
d
e
r

(
S
e
e

N
o
t
e

1
)

1
s
t
-
o
r
d
e
r
,

B
1

i
s

a
p
p
l
i
e
d

t
o

t
h
e

m
e
m
b
e
r

t
o
t
a
l

m
o
m
e
n
t

S
t
r
u
c
t
u
r
e

G
e
o
m
e
t
r
y

U
s
e
d

i
n

t
h
e

A
n
a
l
y
s
i
s

N
o
m
i
n
a
l

(
S
e
e

N
o
t
e

2
)

N
o
m
i
n
a
l

N
o
m
i
n
a
l

N
o
t
i
o
n
a
l

L
o
a
d

t
o

b
e

A
p
p
l
i
e
d

i
n

t
h
e

A
n
a
l
y
s
i
s

0
.
0
0
2
Y
i


M
i
n
i
m
u
m

i
f

2
n
d
/

1
s
t

<

1
.
7


A
d
d
i
t
i
v
e

i
f

2
n
d
/

1
s
t

>

1
.
7

(
S
e
e

N
o
t
e

2
)

0
.
0
0
2
Y
i

m
i
n
i
m
u
m

2
.
1
(

/
L
)
Y
i

>

0
.
0
0
4
2
Y
i

a
d
d
i
t
i
v
e


E
f
f
e
c
t
i
v
e

S
t
i
f
f
n
e
s
s

U
s
e
d

i
n

t
h
e

A
n
a
l
y
s
i
s

0
.
8

*

N
o
m
i
n
a
l
,

e
x
c
e
p
t

E
I
e
f
f

=

0
.
8

b
E
I

w
h
e
n

P
r

>

0
.
5

P
y

(
S
e
e

N
o
t
e

3
)

N
o
m
i
n
a
l

N
o
m
i
n
a
l

b

=

4
[

P
r
/
P
y
(
1

P
r
/
P
y
)
]

U
s
e

o
f

b

=

1

i
s

p
e
r
m
i
t
t
e
d

i
n

a
l
l

c
a
s
e
s

i
f

a
d
d
i
t
i
o
n
a
l

n
o
t
i
o
n
a
l

l
o
a
d
s

o
f

0
.
0
0
1
Y
i

a
r
e

a
p
p
l
i
e
d
,

a
d
d
i
t
i
v
e

t
o

o
t
h
e
r

l
a
t
e
r
a
l

l
o
a
d
s

I
n
-
p
l
a
n
e

f
l
e
x
u
r
a
l


b
u
c
k
l
i
n
g

s
t
r
e
n
g
t
h

P
n
i

P
n
i

i
s

b
a
s
e
d

o
n

t
h
e

u
n
s
u
p
p
o
r
t
e
d

l
e
n
g
t
h

i
n

t
h
e

p
l
a
n
e

o
f

b
e
n
d
i
n
g
,

L
i

(
i
.
e
.
,

K

=

1
)

P
n
i

i
n

m
o
m
e
n
t
-
f
r
a
m
e

c
o
l
u
m
n
s

i
s

b
a
s
e
d

o
n

a

b
u
c
k
l
i
n
g

a
n
a
l
y
s
i
s

o
r

t
h
e

c
o
r
r
e
s
p
o
n
d
i
n
g

e
f
f
e
c
t
i
v
e

l
e
n
g
t
h

K
L
;

P
n
i

i
n

a
l
l

o
t
h
e
r

c
a
s
e
s

i
s

b
a
s
e
d

o
n

K
L
i

=

L
i

(
i
.
e
.
,

K

=

1
)
.


P
n
i

i
s

b
a
s
e
d

o
n

t
h
e

u
n
s
u
p
p
o
r
t
e
d

l
e
n
g
t
h

i
n

t
h
e

p
l
a
n
e

o
f

b
e
n
d
i
n
g
,

L
i
I
f

P
r

<

0
.
1
0

P
e
L
,

o
r

i
f

a

m
e
m
b
e
r

o
u
t
-
o
f
-
s
t
r
a
i
g
h
t
n
e
s
s

o
f

0
.
0
0
1
L

o
r

t
h
e

e
q
u
i
v
a
l
e
n
t

n
o
t
i
o
n
a
l

l
o
a
d
i
n
g

i
s

i
n
c
l
u
d
e
d

i
n

t
h
e

a
n
a
l
y
s
i
s
,

P
n
i

m
a
y

b
e

t
a
k
e
n

e
q
u
a
l

t
o

Q
P
y

(
S
e
e

N
o
t
e
s

4

&

5
)

I
f

2
n
d
/

1
s
t

<

1
.
1
,

K

m
a
y

b
e

t
a
k
e
n

e
q
u
a
l

t
o

o
n
e

i
n

a
l
l

c
a
s
e
s
.

O
u
t
-
o
f
-
p
l
a
n
e

f
l
e
x
u
r
a
l


b
u
c
k
l
i
n
g

s
t
r
e
n
g
t
h

P
n
o

P
n
o

i
s

b
a
s
e
d

o
n

t
h
e

u
n
s
u
p
p
o
r
t
e
d

l
e
n
g
t
h

i
n

t
h
e

o
u
t
-
o
f
-
p
l
a
n
e

d
i
r
e
c
t
i
o
n
,

L
o

A
l
t
e
r
n
a
t
i
v
e
l
y
,

P
n
o

m
a
y

b
e

b
a
s
e
d

o
n

a
n

o
u
t
-
o
f
-
p
l
a
n
e

b
u
c
k
l
i
n
g

a
n
a
l
y
s
i
s

o
r

t
h
e

c
o
r
r
e
s
p
o
n
d
i
n
g

e
f
f
e
c
t
i
v
e

l
e
n
g
t
h

K
L
o









(
s
e
e

N
o
t
e

4
)

36

3
7

N
o
t
e
s

o
n

T
a
b
l
e

2
.
1
:

G
e
n
e
r
a
l

N
o
t
e
.

2
n
d
/

1
s
t

i
s

t
h
e

r
a
t
i
o

o
f

t
h
e

2
n
d
-
o
r
d
e
r

d
r
i
f
t

t
o

t
h
e

1
s
t
-
o
r
d
e
r

d
r
i
f
t

(
f
o
r

r
e
c
t
a
n
g
u
l
a
r

f
r
a
m
e
s
,

2
n
d
/

1
s
t

m
a
y

b
e

t
a
k
e
n

a
s

B
2

c
a
l
c
u
l
a
t
e
d

b
y

S
e
c
t
i
o
n

8
.
2
.
2
)
.

/
L

i
s

t
h
e

l
a
r
g
e
s
t

1
s
t
-
o
r
d
e
r

d
r
i
f
t

f
r
o
m

a
l
l

t
h
e

s
t
o
r
i
e
s

i
n

t
h
e

s
t
r
u
c
t
u
r
e
.


I
n

s
t
r
u
c
t
u
r
e
s

t
h
a
t

h
a
v
e

f
l
e
x
i
b
l
e

d
i
a
p
h
r
a
g
m
s
,

t
h
e

/
L

i
n

e
a
c
h

s
t
o
r
y

i
s

t
a
k
e
n

a
s

t
h
e

a
v
e
r
a
g
e

d
r
i
f
t

w
e
i
g
h
t
e
d

i
n

p
r
o
p
o
r
t
i
o
n

t
o

t
h
e

v
e
r
t
i
c
a
l

l
o
a
d
,

o
r

a
l
t
e
r
n
a
t
i
v
e
l
y
,

t
h
e

m
a
x
i
m
u
m

d
r
i
f
t
.


A
l
l

2
n
d
/

1
s
t

a
n
d

/
L

r
a
t
i
o
s

s
h
a
l
l

b
e

c
a
l
c
u
l
a
t
e
d

u
s
i
n
g

t
h
e

L
R
F
D

l
o
a
d

c
o
m
b
i
n
a
t
i
o
n
s

o
r

u
s
i
n
g

a

f
a
c
t
o
r

o
f

=

1
.
6

a
p
p
l
i
e
d

t
o

t
h
e

g
r
a
v
i
t
y

l
o
a
d
s

i
n

A
S
D
.


T
h
e

f
a
c
t
o
r

i
s

1
.
0

f
o
r

L
R
F
D

a
n
d

1
.
6

f
o
r

A
S
D
.


T
h
e

t
e
r
m

Y
i

i
s

t
h
e

t
o
t
a
l

g
r
a
v
i
t
y

l
o
a
d

a
p
p
l
i
e
d

a
t

a

g
i
v
e
n

l
e
v
e
l

o
f

t
h
e

s
t
r
u
c
t
u
r
e
.


P
e
L

i
s

t
h
e

m
e
m
b
e
r

e
l
a
s
t
i
c

b
u
c
k
l
i
n
g

r
e
s
i
s
t
a
n
c
e

b
a
s
e
d

o
n

t
h
e

a
c
t
u
a
l

u
n
s
u
p
p
o
r
t
e
d

l
e
n
g
t
h

i
n

t
h
e

p
l
a
n
e

o
f

b
e
n
d
i
n
g
,

2
E
I
/
L
2

f
o
r

p
r
i
s
m
a
t
i
c

m
e
m
b
e
r
s
.


N
o
t
e

1
.


A
n
y

l
e
g
i
t
i
m
a
t
e

m
e
t
h
o
d

o
f

s
e
c
o
n
d
-
o
r
d
e
r

a
n
a
l
y
s
i
s

t
h
a
t

i
n
c
l
u
d
e
s

b
o
t
h

P

a
n
d

P

e
f
f
e
c
t
s

i
s

p
e
r
m
i
t
t
e
d
,

i
n
c
l
u
d
i
n
g

1
s
t
-
o
r
d
e
r

a
n
a
l
y
s
i
s

w
i
t
h

a
m
p
l
i
f
i
e
r
s
.


S
e
c
t
i
o
n

C
2
.
1

i
n

A
I
S
C

(
2
0
1
0
)

a
d
d
r
e
s
s
e
s

t
h
e

c
o
n
d
i
t
i
o
n
s

w
h
e
n

t
h
e

P

e
f
f
e
c
t
s

c
a
n

b
e

n
e
g
l
e
c
t
e
d
.


I
n

t
h
e

C
o
m
m
e
n
t
a
r
y

o
f

S
e
c
t
i
o
n

C
2
.
1
,

m
o
r
e

d
e
t
a
i
l
e
d

g
u
i
d
e
l
i
n
e
s

a
r
e

p
r
o
v
i
d
e
d

t
h
e

l
i
m
i
t
s

w
h
e
n

a

P

a
n
a
l
y
s
i
s

c
a
n

a
d
e
q
u
a
t
e
l
y

c
a
p
t
u
r
e

t
h
e

P

e
f
f
e
c
t
s

u
s
i
n
g

o
n
e

e
l
e
m
e
n
t

p
e
r

m
e
m
b
e
r

w
i
t
h

l
e
s
s

t
h
a
n

5

a
n
d

3

%

e
r
r
o
r
s

i
n

t
h
e

s
i
d
e
s
w
a
y

d
i
s
p
l
a
c
e
m
e
n
t
s

a
n
d

t
h
e

c
o
r
r
e
s
p
o
n
d
i
n
g

i
n
t
e
r
n
a
l

m
o
m
e
n
t
s

r
e
s
p
e
c
t
i
v
e
l
y
.


T
h
e

c
a
s
e
s

a
d
d
r
e
s
s
e
d

a
r
e

s
w
a
y

c
o
l
u
m
n
s

w
i
t
h

s
i
m
p
l
y
-
s
u
p
p
o
r
t
e
d

b
a
s
e
d

c
o
n
d
i
t
i
o
n
s
,

s
w
a
y

c
o
l
u
m
n
s

w
i
t
h

r
o
t
a
t
i
o
n
a
l

r
e
s
t
r
a
i
n
t
s

a
t

b
o
t
h

e
n
d
s
,

a
n
d

m
e
m
b
e
r
s

s
u
b
j
e
c
t
e
d

t
o

p
r
e
d
o
m
i
n
a
n
t
l
y

n
o
n
-
s
w
a
y

e
n
d

c
o
n
d
i
t
i
o
n
s
.


I
f

t
h
e

s
p
e
c
i
f
i
e
d

l
i
m
i
t
s

a
r
e

e
x
c
e
e
d
e
d
,

e
i
t
h
e
r

m
u
l
t
i
p
l
e

e
l
e
m
e
n
t
s

m
u
s
t

b
e

u
s
e
d

p
e
r

m
e
m
b
e
r

t
o

o
b
t
a
i
n

a
c
c
u
r
a
t
e

s
e
c
o
n
d
-
o
r
d
e
r

i
n
t
e
r
n
a
l

m
o
m
e
n
t
s

i
n

g
e
n
e
r
a
l

f
r
o
m

a

P
-
l
a
r
g
e

d
e
l
t
a

a
n
a
l
y
s
i
s
,

o
r

a

P
-
s
m
a
l
l

d
e
l
t
a

a
m
p
l
i
f
i
e
r

m
u
s
t

b
e

a
p
p
l
i
e
d

t
o

t
h
e

e
l
e
m
e
n
t

i
n
t
e
r
n
a
l

m
o
m
e
n
t
s
.


K
a
e
h
l
e
r

e
t

a
l
.

(
2
0
1
0
)

p
r
o
v
i
d
e

f
u
r
t
h
e
r

g
u
i
d
e
l
i
n
e
s

f
o
r

t
h
e

a
p
p
r
o
p
r
i
a
t
e

n
u
m
b
e
r

o
f

P

a
n
a
l
y
s
i
s

e
l
e
m
e
n
t
s

t
o

a
d
e
q
u
a
t
e
l
y

c
a
p
t
u
r
e

t
h
e

P
-

e
f
f
e
c
t
s
.

A
c
c
u
r
a
t
e

g
e
n
e
r
a
l

P
-

a
n
a
l
y
s
i
s

s
o
l
u
t
i
o
n
s

m
a
y

b
e

o
b
t
a
i
n
e
d

b
y

m
a
i
n
t
a
i
n
i
n
g

P
r

<

0
.
0
2

P
e
"
,

w
h
e
r
e

P
e
"

i
s

t
h
e

e
l
a
s
t
i
c

b
u
c
k
l
i
n
g

l
o
a
d

b
a
s
e
d

o
n

t
h
e

i
d
e
a
l
i
z
e
d

s
i
m
p
l
y
-
s
u
p
p
o
r
t
e
d

e
n
d

c
o
n
d
i
t
i
o
n
s
,

a
n

e
l
e
m
e
n
t

l
e
n
g
t
h

"
,

a
n
d

E

i
f

t
h
e

a
n
a
l
y
s
i
s

u
s
e
s

t
h
e

n
o
m
i
n
a
l

s
t
i
f
f
n
e
s
s

a
n
d

0
.
8

b
E

i
f

t
h
e

d
i
r
e
c
t

a
n
a
l
y
s
i
s

m
e
t
h
o
d

i
s

u
s
e
d
.





N
o
t
e

2
.


A

n
o
m
i
n
a
l

i
n
i
t
i
a
l

o
u
t
-
o
f
-
p
l
u
m
b
n
e
s
s

o
f

o
/
L

=

0
.
0
0
2

m
a
y

b
e

u
s
e
d

d
i
r
e
c
t
l
y

i
n

l
i
e
u

o
f

a
p
p
l
y
i
n
g

0
.
0
0
2
Y
i

m
i
n
i
m
u
m

o
r

a
d
d
i
t
i
v
e

n
o
t
i
o
n
a
l

l
o
a
d
s
.

2
n
d
/

1
s
t

s
p
e
c
i
f
i
e
d

i
n

t
h
e

d
i
r
e
c
t

a
n
a
l
y
s
i
s

m
e
t
h
o
d

i
s

d
e
t
e
r
m
i
n
e
d

u
s
i
n
g

t
h
e

r
e
d
u
c
e
d

e
f
f
e
c
t
i
v
e

s
t
i
f
f
n
e
s
s

s
p
e
c
i
f
i
e
d

i
n

t
h
e

f
i
r
s
t

c
o
l
u
m
n

o
f

T
a
b
l
e

2
.
1
.

N
o
t
e

3
.


T
h
e

n
o
m
i
n
a
l

s
t
i
f
f
n
e
s
s

a
n
d

g
e
o
m
e
t
r
y

s
h
o
u
l
d

b
e

e
m
p
l
o
y
e
d

f
o
r

c
h
e
c
k
i
n
g

s
e
r
v
i
c
e
a
b
i
l
i
t
y

l
i
m
i
t

s
t
a
t
e
s
.


T
h
e

r
e
d
u
c
e
d

e
f
f
e
c
t
i
v
e

s
t
i
f
f
n
e
s
s

a
n
d

t
h
e

n
o
t
i
o
n
a
l

l
o
a
d
s

o
r

n
o
m
i
n
a
l

i
n
i
t
i
a
l

o
u
t
-
o
f
-
p
l
u
m
b
n
e
s
s

a
r
e

r
e
q
u
i
r
e
d

o
n
l
y

i
n

c
o
n
s
i
d
e
r
i
n
g

s
t
r
e
n
g
t
h

l
i
m
i
t

s
t
a
t
e
s
.


N
o
t
e

4
.


A
I
S
C

(
2
0
1
0
)

d
o
e
s

n
o
t

e
x
p
l
i
c
i
t
l
y

s
t
a
t
e

t
h
i
s

p
r
o
v
i
s
i
o
n

i
n

t
h
e

c
o
n
t
e
x
t

o
f

t
h
e

d
i
r
e
c
t

a
n
a
l
y
s
i
s

m
e
t
h
o
d
.


T
h
i
s

p
r
o
v
i
s
i
o
n

i
s

e
n
c
o
m
p
a
s
s
e
d

w
i
t
h
i
n

t
h
e

C
h
a
p
t
e
r

C

r
e
q
u
i
r
e
m
e
n
t
s

f
o
r

g
e
n
e
r
a
l

s
t
a
b
i
l
i
t
y

a
n
a
l
y
s
i
s

a
n
d

d
e
s
i
g
n
,

w
h
i
c
h

a
l
l
o
w

a
n
y

m
e
t
h
o
d

o
f

a
n
a
l
y
s
i
s

a
n
d

d
e
s
i
g
n

t
h
a
t

a
d
d
r
e
s
s
e
s

t
h
e

e
f
f
e
c
t
s

l
i
s
t
e
d

a
t

t
h
e

b
e
g
i
n
n
i
n
g

o
f

S
e
c
t
i
o
n

2
.
1
.



N
o
t
e

5
.


T
h
e

l
a
r
g
e
s
t

u
n
c
o
n
s
e
r
v
a
t
i
v
e

e
r
r
o
r

a
s
s
o
c
i
a
t
e
d

w
i
t
h

t
h
e

l
i
m
i
t

P
r

<

0
.
1
0

P
e
L

i
s

a
p
p
r
o
x
i
m
a
t
e
l
y

f
i
v
e

p
e
r
c
e
n
t

a
n
d

o
c
c
u
r
s

f
o
r

a

s
i
m
p
l
y
-
s
u
p
p
o
r
t
e
d
,

c
o
n
c
e
n
t
r
i
c
a
l
l
y

l
o
a
d
e
d

c
o
l
u
m
n

w
i
t
h

z
e
r
o

m
o
m
e
n
t
,

Q

=

1
,

a
n
d

P
r

=

0
.
1
0

P
e
L

=

c
P
y
.


T
h
e

t
a
r
g
e
t

o
f

f
i
v
e

p
e
r
c
e
n
t

m
a
x
i
m
u
m

u
n
c
o
n
s
e
r
v
a
t
i
v
e

e
r
r
o
r

i
s

b
a
s
e
d

o
n

t
h
e

o
r
i
g
i
n
a
l

d
e
v
e
l
o
p
m
e
n
t

o
f

t
h
e

A
I
S
C

L
R
F
D

b
e
a
m
-
c
o
l
u
m
n

s
t
r
e
n
g
t
h

e
q
u
a
t
i
o
n
s

(
A
S
C
E

1
9
9
7
;

S
u
r
o
v
e
k
-
M
a
l
e
c
k

a
n
d

W
h
i
t
e

2
0
0
4
a
)
.

N
o
t
e

6
.


T
h
e

1
s
t
-
o
r
d
e
r

a
n
a
l
y
s
i
s

m
e
t
h
o
d

d
o
e
s

n
o
t

a
c
c
o
u
n
t

f
o
r

t
h
e

i
n
f
l
u
e
n
c
e

o
f

s
i
g
n
i
f
i
c
a
n
t

a
x
i
a
l

c
o
m
p
r
e
s
s
i
o
n

i
n

t
h
e

r
a
f
t
e
r
s

o
f

c
l
e
a
r
-
s
p
a
n

f
r
a
m
e
s
.


T
h
e
r
e
f
o
r
e
,

t
h
i
s

m
e
t
h
o
d

s
t
r
i
c
t
l
y

s
h
o
u
l
d

n
o
t

b
e

a
p
p
l
i
e
d

f
o
r

t
h
e

a
n
a
l
y
s
i
s

a
n
d

d
e
s
i
g
n

o
f

t
h
e

p
r
i
m
a
r
y

m
o
m
e
n
t

f
r
a
m
e
s

i
n

t
h
e
s
e

t
y
p
e
s

o
f

s
t
r
u
c
t
u
r
e
s
.


37

38

inelasticity and reliability considerations at the strength limit of the most critical member
or members, as well as (2) the effects of a nominal initial out-of-plumbness
o
, within
fabrication and erection tolerances, on the internal forces and moments at the above
strength limit. The direct analysis method provides an improved representation of the
actual second-order inelastic forces and moments in the structure at the strength limit of
the most critical member or members. Due to this improvement in the calculation of the
internal forces and moments, AISC (2010) bases its calculation of P
ni
, the column
nominal strength for checking the in-plane resistance in Eqs. 2.5, on the actual
unsupported length in the plane of bending.
Interestingly, the use of the stub-column strength for P
ni
(QP
y
for columns with
slender compression elements) was actively considered in the development of the direct
analysis approach (Surovek-Maleck and White 2004a). Although this is a viable option,
it requires the modeling of out-of-straightness in the analysis for members subjected to
large axial compression (to properly capture in-plane limit states dominated by non-sway
column flexural-buckling). The modeling of member out-of-straightness adds an
additional level of complexity to the analysis, and in many steel structures, P
ni
based on
the actual unsupported length is only slightly smaller than QP
y
. Therefore, AISC (2010)
recommends the use of P
ni
based on the actual unsupported length. However, in many
metal building structural systems, the member axial loads are small enough such that the
beam-column resistance is represented accurately using P
ni
= QP
y
, without the inclusion
of any member out-of-straightness in the analysis. In other cases the modeling of
member out-of-straightness may be needed to accurately capture the strength limit.

39

2.3 Effective Length Method
The effective length method is in essence the traditional AISC method of design
since 1961, but with the addition of a notional minimum lateral load for gravity-load only
combinations. This minimum lateral load accounts for the influence of nominal
geometric imperfections on the brace forces, beam moments, connection moments and
in-plane member moments used for out-of-plane strength design of beam-columns. In
actuality, the effects of any physical out-of-plumbness exist for all load combinations.
However, these effects tend to be small and are overwhelmed by the effects of the
primary lateral loads in all the ASCE 7 lateral load combinations, as long as the
structures sidesway amplification is not excessive. Therefore, in the AISC (2005)
effective length method, the notional lateral loads are specified solely as minimum lateral
loads in the gravity load only combinations.
AISC (2010) disallows the use of the effective length method when the second-
order amplification of the sidesway displacements is larger than 1.7, i.e.,
2nd
/
1st
> 1.7
(based on the nominal elastic stiffness of the structure). This is due to the fact that the
effective length method significantly underestimates the internal forces and moments in
certain cases when this limit is exceeded (Deierlein 2003 & 2004; Kuchenbecker et al.
2004; White et al. 2006). For structures with
2nd
/
1st
> 1.7, AISC (2010) requires the
use of the direct analysis method. Correspondingly, when using the direct analysis
approach with structures having
2nd
/
1st
< 1.7, AISC (2005) allows the Engineer to
apply the notional lateral loads (or the corresponding nominal out-of-plumbness) as
minimum values used solely with the gravity load only combinations.
40

For column and beam-column in-plane strength assessment in moment frames,
the effective length approach generally focuses on the calculation of the member axial
stresses F
ei
at incipient buckling of an appropriately selected model (the subscript i is
used to denote in-plane flexural buckling). This buckling model is usually some type of
subassembly that is isolated from the rest of the structural system (ASCE 1997).
Engineers often handle the elastic buckling stresses (F
ei
) implicitly, via the corresponding
column effective lengths KL
i
. The effective length is related to the underlying elastic
buckling stress via the relationship
2
2
) / (
i i
ei
r KL
E
F

= (Eq. 2.6a)
or
ei
i i
i
F
r L E
K
2 2
) / /(
= (Eq. 2.6b)
In the effective length method, the influences of residual stresses, P-
o
effects and P-
o

effects are addressed implicitly in the calculation of P
ni
from the column strength
equations. These equations can be written either in terms of KL
i
or F
ei
(AISC 2010).
Unfortunately, the selection of an appropriate subassembly buckling model generally
requires considerable skill and engineering judgment. As a result, there is a plethora of
different buckling models and K factor calculations. In certain cases the different models
can produce radically different results. A few examples are provided below
In particular, one should note that a rigorous buckling analysis of the complete
structure does not in general provide an appropriate F
ei
or K
i
. Members that have small
axial stress F
ei
at the buckling limit (relative to
2
E/(L
i
/r
i
)
2
) tend to have high values for
41

K
i
in Eq. 2.6b. In some cases, these large K
i
values are justified while in other cases they
are not. If the member is indeed participating in the governing buckling mode, a large K
i

is justified. If the member is largely undergoing rigid-body motion in the governing
buckling mode, or if it has a relatively light axial load and is predominantly serving to
restrain the buckling of other members, a large K
i
value is sometimes not justified. The
distinction between these two situations requires engineering judgment. Furthermore, the
concept of effective length is more obscure and less useful for general nonprismatic
members subjected to nonuniform axial compression.
Some of the situations requiring the greatest exercise of judgment to avoid
excessively large K values include: (1) columns in the upper stories of tall buildings, (2)
columns with highly flexible and/or weak connections and (3) beams or rafters in portal
frames, which may have significant axial compression due to the horizontal thrust at the
base of the frame. There is no simple way of quantifying the relative participation of a
given member in the overall buckling of the structure or subassembly under
consideration. Quantifying the participation requires an analysis of the sensitivity of the
buckling load to variations in the member sizes. Even if one conducted such an analysis,
there is no established metric for judging when Eq. 2.6b should or should not be used.
Engineers typically base their effective length calculations on story-by-story models to
avoid the first of the above situations. They idealize columns with weak and/or flexible
connections as pin-ended leaner columns with K = 1 to avoid the second situation.
Lastly, they often use K = 1, or K < 1 (counting on rotational restraint from the sidesway
columns), for design of the beams or rafters in portal frames, although the F
ei
of these
42

members obtained from an eigenvalue buckling analysis of the full structure may suggest
K > 1 via Eq. 2.6b.
The direct analysis method provides a simpler more accurate way of addressing
frame in-plane stability considerations. By including an appropriately reduced nominal
elastic stiffness, an appropriate nominal out-of-plumbness of the structure, and an
appropriate out-of-straightness (for members subjected to high axial loads) in the
analysis, the member length considerations can be completely removed from the
resistance side of the design equations. The member in-plane column strength term P
ni
is
simply taken equal to QP
y
. In-plane stability is addressed by estimating the required
internal cross-section strengths P
r
and M
r
directly from the analysis, and by comparing
these required strengths against appropriate cross-section resistances.
2.4 First-Order Analysis Method
The first-order analysis method, summarized in Table 2.1, is in essence a
simplified conservative application of the direct analysis approach, targeted at rectangular
or tiered building frames. This method involves:
The implicit application of a conservative sidesway amplification factor of 1.5
(conservative as long as
2nd
/
1st
< 1.5) to the 1
st
-order story drift /L or a nominal
initial out-of-plumbness of
o
/L = 0.002, whichever is larger. The 1
st
-order story drift
/L is taken as the largest drift from all the stories in the structure, calculated under
the LRFD load combinations or with a factor of 1.6 applied to the gravity loads in
ASD. In structures that have flexible diaphragms, the /L in each story is taken as the
average drift weighted in proportion to the vertical load, or alternatively, the
maximum drift
43

The inclusion of the direct analysis stiffness reduction factor of 0.8 implicitly in the
second-order amplification of the above /L or
o
/L, resulting in an amplification of
the sidesway displacements by a factor of 2.1 rather than 1.5.
The assumption that all the stories of the structure have a sidesway deflection equal to
the above maximum amplified value.
Inclusion of the corresponding P- effects in the 1
st
-order analysis, by applying the P-
shear forces corresponding to the above sidesway displacement at each level of the
structure (these P- shears are applied in addition to any other lateral loads).
Amplification of the corresponding total member moments obtained from the analysis
by the non-sway amplification factor B
1
calculated as specified in Appendix 8 in
AISC (2010).
The first-order analysis method is restricted to frames with
2nd-order
/
1st-order
< 1.5 as well
as to cases where P
r
< 0.5P
y
in all of the members whose flexural stiffnesses contribute
to the lateral stability of the structure. The limit P
r
< 0.5P
y
prevents the application of
the method to structures where the sidesway stiffnesses are reduced significantly by
combined residual stress effects and large column axial compression.
Although the

first-order analysis method can be useful for simplified analysis and
design of metal buildings in their longitudinal braced direction, this method does not
include the effects of rafter axial compression on the flexural response of the primary
moment frames. Also, this method is really just a direct analysis with a number of
simplifying assumptions. There are numerous other ways to apply direct analysis using
an approximate second-order analysis, many of which are less conservative than the
44

above approach. Therefore, the first-order analysis method is not considered further in
this research.
It should be noted that both the direct analysis method and effective length
method require a second-order elastic analysis. However, any legitimate method of
second-order elastic analysis is allowed, including first-order analysis with amplifiers,
when the amplifiers are sufficiently accurate. The stiffnesses and notional lateral loads
(or nominal geometric imperfections) used in the analysis are different in each of the
methods (see Table 2.1).
The beam-column out-of-plane resistance check is the same in both of the above
methods, albeit with different values of P
r
and M
r
. In AISC (2010), the simplest out-of-
plane beam-column resistance check is given by Eqs. 2.5 but with P
n
= P
no
, where P
no
is
the out-of-plane flexural or flexural-torsional buckling strength of the member as a
concentrically-loaded column. Other enhanced beam-column out-of-plane strength
checks are provided in AISC (2010) and are discussed in Sections 2.6.2 and 3.4.1.
2.5 Fundamental Comparison of The Direct Analysis and
Effective Length Methods
The differences between the direct analysis and the effective length methods are
predominantly in the way that they handle the beam-column in-plane strength check.
Figures 2.6a and 2.6b, adapted from Deierlein (2004), illustrate the fundamental
differences. Figure 2.6a shows a representative beam-column in-plane check using the
traditional effective length approach, i.e., the effective length method as outlined in Table
2.1 but with no limitations on the use of the method and with zero notional load. The
dashed curve in the figure is the AISC (2010) beam-column strength envelope, given by
Eqs. 2.5. The anchor point for this curve on the vertical axis, P
n(KLi)
, is the member
45

nominal axial strength determined using the effective length KL
i
(or equivalently, using
the member elastic buckling stress F
ei
as illustrated by Eqs. 2.6). The anchor point on the
horizontal axis is the member in-plane flexural resistance M
ni
, which is based in AISC
(2010) either on flange local buckling or general flexural yielding considerations. The
other two curves in the plot indicate the member internal axial force and moment under
increasing applied loads on the structure. The curve labeled actual response is
determined from an experiment or from a rigorous second-order distributed plasticity
analysis that accounts for all the significant stability effects, whereas the second curve is
from a second-order elastic analysis of the idealized straight and plumb, nominally elastic
structure. The actual response curve indicates larger moment than the second-order
elastic analysis curve due to the combined effects of partial yielding and geometric
imperfections, which are not included in the elastic second-order analysis. The maximum
value of P on the actual response curve (P
max
) is the largest value of the axial force that
the member can sustain at its stability limit. Correspondingly, the nominal design
strength is defined by the intersection of the force-point trace from the second-order
elastic analysis with the P
n(KLi)
based envelope. The effective length provisions have
been calibrated such that this intersection point gives an accurate to conservative estimate
of the actual maximum strength.
The reduced stiffness and the notional load (or the corresponding nominal out-of-
plumbness) in the direct analysis method are calibrated to estimate the actual response
using a second-order elastic analysis. This is illustrated by Figure 2.6b, where the force-
point trace of the member internal axial force and moment from the second-order elastic
46


Figure 2.6. Comparison of beam-column strength interaction checks for (a) the effective
length method (with zero notional load) and (b) the direct analysis method.
analysis (conducted using the reduced stiffness and notional lateral loads) is close to the
actual response plot. The calibration is done to achieve parity between the actual
compressive strength (indicated by P
max
on the actual response curve) and the nominal
design strength. By accounting for residual stress, partial member yielding and geometric
imperfection effects in the second-order elastic analysis, the resistance can be checked on
a member cross-section basis. That is, the anchor point on the vertical axis for the beam-
column strength envelope can be taken as QP
y
. As noted in Section 2.1, the use of P
ni
=
QP
y
requires that the member axial force must be smaller than 0.10P
eL
, or the use of a
member out-of-straightness of 0.001L in the analysis (to capture in-plane limit states
dominated by non-sway column flexural-buckling). AISC (2010) bases P
ni
on the actual
unsupported length in the plane of bending to avoid the need for consideration of member
out-of-straightness in a general frame analysis. The anchor point on the horizontal axis
(M
ni
) is the same in both the direct analysis and the effective length methods. In the
47

direct analysis method, the internal force and moment (P and M) and the strength
envelope (with the anchor points QP
y
and M
ni
) are an improved representation of the
actual response. Conversely, in the traditional effective length approach (i.e., the
effective length method with no notional load included), smaller idealized values of P
and M are checked against a correspondingly reduced beam-column strength interaction
curve. The direct analysis method accounts for the in-plane system stability effects
directly within the second-order analysis. Conversely, the effective length method
accounts for the in-plane system stability effects by reducing the member axial strength
P
ni
via an effective length KL
i
or the elastic buckling stress F
ei
obtained (implicitly or
explicitly) from an appropriately configured buckling analysis.
2.6 Illustrative Examples
The concepts discussed in Section 2.5 are best understood by considering a few
simple examples. The following subsections highlight two basic case studies, one taken
from the AISC (2005) TC10 and Specification Committee developments and the other
created by modifying an example design problem considered by the Specification
Committee. Other more detailed examples are presented by Maleck (2001), Maritinez-
Garcia (2002), Deierlein (2003), Surovek-Maleck and White (2004a & b), Kuchenbecker
et al. (2004), and White et al. (2006 and 2007a and b).
2.6.1 Cantilever Beam-Column
One of the simplest illustrations of the direct analysis and effective length
methods is the solution for the design strength of a fixed-base cantilever composed of a
rolled wide-flange section. Figure 2.7a shows a W10x60 cantilever subjected to a
vertical load P and a proportional horizontal load of H = 0.01P, adapted from Deierlein
48

(2004). The bending is about the major axis and the member is braced out-of-plane such
that its in-plane resistance governs. In this problem, the member in-plane strength
governs in both the direct analysis and the effective length methods if the member is
braced at its top and bottom in the out-of-plane direction, the enhanced AISC (2005)
beam-column strength interaction equations are employed, and K = 0.7 is used for the
calculation of P
no
, the column strength in the out-of-plane direction. The column
slenderness in the plane of bending is L/r

= 40 based on the members actual length and
KL/r = 80 based on the effective length (with K = 2).
Figure 2.7b shows plots of the axial load versus the moment at the column base,
determined using three approaches: (1) the traditional effective length method (no
minimum notional lateral load included), (2) the direct analysis method, and (3) a
rigorous second-order distributed plasticity analysis. Load and Resistance Factor Design
(LRFD) is used with a resistance factor of
c
=
b
= 0.9 in each of these solutions. The
rigorous distributed plasticity analysis is based on a factored stiffness and strength of
0.9E and 0.9F
y
, an out-of-plumbness and out-of-straightness on the geometry of 0.002L
and 0.001L respectively (oriented in the same direction as the bending due to the applied
loads), the Lehigh residual stress pattern (Galambos and Ketter 1959) with a maximum
compressive nominal residual stress at the flange tips of 0.3(0.9F
y
) = 0.27F
y
, and an
assumed elastic-perfectly plastic material stress-strain response. A small post-yield
stiffness of 0.001E is used for numerical stability purposes. These are established
parameters for calculation of benchmark design strengths in LRFD using a distributed
plasticity analysis (ASCE 1997; Martinez-Garcia 2002; Deierlein 2003; Maleck and
White 2003; Surovek-Maleck and White 2004b; White et al. 2006). AISC (2010)
49

0
100
200
300
400
500
600
700
800
0 50 100 150 200 250 300
Effective Length
Direct Analysis
W10x60
F
y
= 50 ksi
L/r
x
= 40
14'-8"
P
r
H = 0.01P
r
(a) (b)
Distributed Plasticity
Analysis
c
P
y
M
r
(ft-kips)
P
eLx
= 3151 kips
P
r


(
k
i
p
s
)
0.10P
eLx
= 315 kips
c
P
n(KLi)

Figure 2.7. Cantilever beam-column example.
specified these analysis requirements in Appendix 1, Design by Inelastic Analysis.
One should note that the internal moment from the direct analysis is larger than
that obtained from the traditional effective length method. This is due to the use of a
reduced stiffness of 0.8EI
x
, as well as a notional lateral load of 0.002P, which is added to
the applied lateral load. Although the axial load at the intersection of the force-point
trace with the strength envelope is slightly larger than 0.10P
eLx
= 0.10(
2
EI
x
/L
2
) = 315
kips in this problem, the member out-of-straightness effects are not considered in the
second-order elastic analysis by the direct analysis method and the axial load anchor
point for the direct analysis strength curve is taken as
c
P
y
= 796 kips. The plot in Figure
2.7b shows that the direct analysis internal moments are very similar to the internal
moments calculated by the rigorous inelastic analysis. Overlaid on the above force-point
traces are the beam-column strength envelopes, where the
c
P
n
anchor points are
c
P
n(KLi)
= 496 kips for the effective length method and
c
P
y
= 796 kips for the direct analysis
50

method.
The design strengths, determined as the combined P and M at the intersections
with the in-plane beam-column strength curves (Eqs. 2.5), are summarized in Table 2.2.
The ratios of the maximum base moments M
max
= HL + P( +
o
) to the primary moment
HL indicate the magnitude of the second-order effects. The axial load at the direct
analysis strength limit, which is representative of the strength in terms of the total applied
load, is five percent higher than that obtained from the distributed plasticity analysis.
Conversely, the axial load at the effective length method beam-column strength limit is
four percent smaller than that obtained from the distributed plasticity solution. Both of
these estimates are within the targeted upper bound of five percent unconservative error
relative to the refined analysis established in the original development of the AISC LRFD
beam-column strength equations (ASCE 1997; Surovek-Maleck and White 2004a).
Table 2.2. Summary of calculated design strengths, cantilever beam-column example.
W

,> W

W
W

u A
u A
1 L L

u M


The difference in the calculated internal moments is much larger. This difference
is expected since the effective length approach compensates for the underestimation of
the actual moments by reducing the value of the axial resistance term P
ni
, whereas the
direct analysis method imposes additional requirements on the analysis to obtain an
51

improved estimate of the actual internal moments. This more accurate calculation of the
internal moments also influences the design of the restraining members and their
connections. For instance, in this example, the column base moments from the direct
analysis method are more representative of the actual moments required at this position to
support the applied loads associated with the calculated member resistance. In this
regard, the direct analysis method provides a direct calculation of the required strengths
for all of the structural components. Conversely, the traditional effective length approach
generally necessitates supplementary requirements for calculation of the required
component strengths. AISC (2010) implements these supplementary requirements as (1)
a minimum notional lateral load to be applied with gravity-only load combinations and
(2) a limit on the use of the effective length method to frames having
2nd
/
1st
< 1.5, as
summarized in Table 2.1.
2.6.2 Single Story Rectangular Frame
Figure 2.8 shows a slightly modified version of one of the design problems posed
during the process of validating and checking the final AISC (2005) provisions by the
Specification Committee (DP-13). This problem is a single-story rectangular frame. It
provides a somewhat more realistic illustration of the potential of AISC (2005 and 2010)
for certain types of metal building frames, although it uses rolled I-section members. The
interior columns in the structure are all leaning columns. All the beam-to-column
connections are simple except for the connections to the exterior columns, which are
assumed to be fully-restrained. The beams are made continuous over the interior
columns in the modified example, whereas the frame studied by the AISC Specification
Committee used simple connections at the ends of the beams for all the interior joints.
52

This modification is performed to simulate typical conditions at the interior columns in
modular metal building frames. Due to the continuity of the beams, the exterior columns
are reduced from W12x72 sections in the AISC frame to W12x65 sections in the
modified example. Also, the beams are reduced from W24x68 sections in the exterior
spans and W27x84 sections in the interior spans of the AISC frame to W24x62 sections
in this example. The resulting modified frame has similar drift characteristics under
lateral loads; both the AISC example and the modified example satisfy a maximum drift
criterion of L/100 for the nominal (unfactored) wind load based on a first-order elastic
analysis. The strength behavior of the lateral load resisting beams and columns is similar
in both frames, with of course the exception of the beam continuity effects over the
interior columns in the modified design. The reader is referred to White et al. (2006) for a
detailed discussion of the behavior of the AISC frame.

Figure 2.8. Modified version of AISC single-story rectangular frame example DP-13.
53

The exterior columns in both examples are subjected to relatively light axial
loads, whereas they experience substantial gravity load moments as well as significant
wind load moments. Also, the frames have significant second-order effects, i.e.,
amplification of the member internal bending moments. The columns are braced in the
out-of-plane direction at their base and at the roof height. Simple base conditions are
assumed in both the in-plane and out-of-plane directions, and simple connections are
assumed in the out-of-plane direction at the column tops. The beams are assumed to be
braced sufficiently such that their flexural resistance is equal to M
p
. This problem is
considered for the following LRFD load combinations:
1. Load Case 1 (LC1): 1.2 Dead + 1.6 Snow
2. Load Case 2 (LC2): 1.2 Dead + 0.5 Snow + 1.6 Wind
Figure 2.9 shows the applied fraction of the design loads versus the story drift for
these two load combinations on the modified single-story frame, obtained from a rigorous
second-order distributed plasticity analysis having the same attributes as described above
for the cantilever beam-column. The distributed plasticity analysis gives a maximum in-
plane capacity of 1.19 times the factored design load level for LC1 and 1.03 of the
factored design load level for LC2. Table 2.3 compares the fractions of the design loads
giving a unity check of 1.0 for the in-plane strength from Eqs. 2.5 for each of the design
methods to the above load capacities from the distributed plasticity analysis. In this
frame, the right-hand side exterior column is the most critical member in the effective
length check for both load cases and in the direct analysis check for LC2. In the direct
analysis check for LC1, the negative beam moment over the left-most interior column
gives the most critical strength check. One can observe that the direct analysis
54

F
r
a
c
t
i
o
n

o
f

D
e
s
i
g
n

L
o
a
d

Figure 2.9. Load versus story drift from distributed plasticity analysis for load cases 1
and 2, modified DP-13 example.
Table 2.3. Fraction of design loads corresponding to a unity check of 1.0 for the right-
hand beam-column, and maximum capacities predicted by distributed plasticity analysis,
modified DP-13 example.
u M LC LC
u A
u A
1 L L



method closely captures the resistances from the refined distributed plasticity solution,
predicting that 1.18 of LC1 and 0.95 of LC2 can be applied prior to reaching the
maximum strength of the critical leeward column. That is, the strength for the gravity
load combination is underestimated by 0.8 percent, while the strength for the wind load
combination is underestimated by 7.8 percent. The larger underestimation for the wind
load case is largely due to inelastic redistribution in the structure after a plastic hinge
55

forms at the top of the leeward column. However, there is little redundancy in the
structural system as the column strength limit is approached for the gravity load
combination. Nevertheless, the most critical member check for LC1 by direct analysis is
the beam negative moment check over the left-most interior column. The large elastic
negative bending moment at this location reaches the factored plastic moment resistance
of the beam
b
M
p
at 1.10 of the design load. The distributed plasticity solution for LC1
indicates substantial inelastic redistribution from a beam plastic hinge at this location
prior to the structure reaching its strength limit at 1.19 of the design load. The right-most
exterior column is critical for all the other design checks by either the direct analysis or
the effective length methods.
The traditional effective length method (with zero notional lateral load)
dramatically underestimates the resistance of the modified DP-13 frame, satisfying the
unity check of Eqs. 2.5 at 1.05 and 0.81 of the design load levels for LC1 and LC2
respectively. That is, the effective length method gives a capacity of only 1.05/1.19 =
0.88 and 0.81/1.03 = 0.79 of the in-plane capacity from the distributed plasticity analysis
for LC1 and LC2. A key reason for these underestimations is illustrated in Figures 2.10
and 2.11, which show the force-point traces for the axial load and moment in the leeward
beam-column by each of the solutions versus the in-plane beam-column resistance
envelopes for LC1 and LC2 respectively. The significant in-plane stability effects in this
example result in a P
n(KLi)
of only 0.11P
y
for LC1 and 0.09P
y
for LC2. The
corresponding effective length factors for the leeward column are 5.11 and 4.75 using a
rigorous sidesway buckling analysis for each of the load combinations. The direct
analysis method gives a substantially larger estimate of the in-plane resistance because it
56

focuses on a more realistic estimate of the internal moments and the corresponding
member resistances. In determining the anchor point P
n(KLi)
for its beam-column strength
envelope, the effective length method overemphasizes the response of the structure to
unrealistic loads causing uniform axial compression in all of its columns. The physical
strength of this frame is dominated by the amplified internal moments (a large fraction of
which are not related to the sidesway of the structure) reaching the flexural resistance of
the leeward beam-column, not by a column failure under concentrically-applied member
axial loads.
0
20
40
60
80
100
0 100 200 300 400
Effective Length
Direct Analysis
Distributed Plasticity
Analysis
c
P
y
= 860 kips
M
r
(ft-kips)
c
P
n(KLi)

Figure 2.10. Force-point trace for the leeward beam-column by the effective length
method, the direct analysis method and distributed plasticity analysis, LC1, modified DP-
13 example.
57

0
20
40
60
80
100
0 100 200 300 400
Effective Length
Direct Analysis
Distributed Plasticity
Analysis
c
P
y
= 860 kips
M
r
(ft-kips)
P
r


(
k
i
p
s
)
c
P
n(KLi)

Figure 2.11. Force-point trace for the leeward beam-column by the effective length
method, the direct analysis method and distributed plasticity analysis, LC2, modified DP-
13 example.
AISC (2010) provides the following enhanced beam-column strength interaction
equation for checking the out-of-plane resistance of doubly-symmetric rolled I-section
members subjected to single axis flexure and compression:
0 . 1 5 . 0 5 . 1
2
) 1 (

= Cb cx b
r
co
r
co
r
M C
M
P
P
P
P
(Eq. 2.7, AISC H1-2)
where P
co
is the out-of-plane column strength, M
cx
is the governing major-axis flexural
resistance of the member determined in accordance with Chapter F using C
b
= 1, and C
b

is lateral-torsional buckling modification factor specified in Section F1. It should be
emphasized that the value of C
b
M
cx(Cb=1)
may be larger than
b
M
px
in LRFD or M
px
/
b
in
ASD. Equation 2.7 is derived from the fundamental form for the out-of-plane lateral-
torsional buckling strength of doubly-symmetric I-section members in LRFD,
58

= ez c
r
no c
r
Cb nx b b
r
P
P
P
P
M C
M
1 1
2
) 1 (
(Eq. 2.8, AISC C-H1-6)
where P
ez
is the elastic torsional buckling strength. Equation 2.7 is developed by
substituting a lower bound of 2.0 for the ratio of P
ez
/P
no
for W section members with KL
o

= KL
z
. This enhanced out-of-plane beam-column strength interaction equation in AISC
(2010) is improved from the one in AISC (2005), which assumes the upper bound of
P
ez
/P
no
= . In addition, in the prior AISC (2005) provisions, the fact that C
b
M
nx(Cb=1)

may be larger than M
p
was not apparent. Based on Eq. 2.7, the out-of-plane resistance
does not govern for both LC1 and LC2 for the above example frame.